Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Association among Physical Activity Level, Cardiorespiratory Fitness, and Risk of Musculoskeletal Injury

Association among Physical Activity Level, Cardiorespiratory Fitness, and Risk of Musculoskeletal... Abstract To help public health practitioners promote physical activities with a low risk of injury, this study determined the relation among type and duration of physical activity, cardiorespiratory fitness, and musculoskeletal injury in a sample of adults enrolled in the Aerobics Center Longitudinal Study. Subjects included 4,034 men and 967 women who underwent a baseline physical examination between 1970 and 1985 and who returned a mailed follow-up survey in 1986. At baseline, a treadmill graded exercise test was used to measure cardiorespiratory fitness. At follow-up, subjects reported injuries and type and duration of physical activity in the preceding 12 months. Polytomous logistic regression was used to estimate the association among physical activity type and duration, cardiorespiratory fitness, and injury. The risk of sustaining an activity-related injury increased with higher duration of physical activity per week and cardiorespiratory fitness levels. Results suggest that cardiorespiratory fitness may be a surrogate for unmeasured components of physical activity, such as exercise intensity. Among walkers, increasing duration of activity per week was not associated with an increased risk of injury. Results suggest that, for most adults, walking is a safe form of physical activity associated with a lower risk of injury than running or sport participation. exercise, logistic models, musculoskeletal system, physical fitness, wounds and injuries Physical inactivity is a leading public health problem associated with decreased longevity as well as cardiovascular disease, cancer, obesity, diabetes, and other diseases (1–3). Recently, private and federal organizations have published physical activity recommendations to increase physical activity levels in the US general adult population (3–5). In a joint statement in 1995, the Centers for Disease Control and Prevention and the American College of Sports Medicine recommended that all adults accumulate bouts of moderate physical activity totaling at least 30 minutes on most, if not all, days of the week (5). Walking, jogging, tennis, volleyball, bicycling, and swimming are popular activities that meet these recommendations (3, 5). The risk of injury is one of the few adverse effects of a physically active lifestyle (3, 5–8). Among vigorously active populations, a clear dose-response relation exists between weight-bearing exercise volume and risk of lower extremity injury (7, 9–19). Studies of runners and military trainees have found that higher levels of cardiorespiratory fitness are associated with lower risk of exercise-related musculoskeletal injury (13, 15, 20, 21). However, the relation among physical activity, cardiorespiratory fitness, and musculoskeletal injury has not been defined for the civilian general population that participates in the moderate types of exercise recommended by national health organizations, such as walking, jogging, or sport activities. Identifying the relation between different levels of physical activity and musculoskeletal injury may help public health practitioners promote safe physical activities for adults. The purpose of this study was to identify the association among physical activity type and duration, cardiorespiratory fitness, and musculoskeletal injury in a group of adults with various physical activity and cardiorespiratory fitness levels. MATERIALS AND METHODS Subjects The data for this study were obtained from the Aerobics Center Longitudinal Study, a prospective epidemiologic study investigating the influence of physical activity and cardiorespiratory fitness on health outcomes among healthy adults. Potential subjects attended the Cooper Clinic, a preventive medicine clinic in Dallas, Texas, where they underwent a baseline physical examination, between 1970 and 1981, that included a personal and family health history questionnaire, a physician examination, anthropometric measurements, blood analyses, and a treadmill graded exercise test. Subjects were enrolled in the Aerobics Center Longitudinal Study after reading and signing a consent form. The consent form and the Aerobics Center Longitudinal Study protocol are reviewed annually by the Cooper Institute for Aerobics Research review board. Upon enrollment, the health status of subjects was monitored approximately every 5 years by mailed follow-up surveys. Of the 18,806 baseline clinical examinations through 1981, only subjects (n = 11,972) that returned a follow-up survey in 1982 were mailed the 1986 survey. A total of 5,028 men and 1,283 women returned the 1986 survey. After deceased subjects and incorrect addresses were taken into account, the response rate of the 1986 survey was 72 percent. Subjects eligible for this study were adults aged 20–85 years (at follow-up) who completed the baseline examination and returned a follow-up survey in 1986. Analyses were limited to the 4,034 men and 967 women who returned the survey and had complete data for all study variables. Additional details of the Aerobics Center Longitudinal Study methods have been described previously (1, 2). Injury definition An injury was any self-reported muscle, tendon, bone, ligament, or joint injury occurring in the 12 months before the mailed follow-up survey. The cause of injury was defined as “activity related” if injury was due to participation in a run/walk/jog or strenuous sport exercise program (e.g., racquet sports, cycling, swimming, aerobic dance, basketball, soccer, and so on). It was defined as “other injury” if the injury was due to a cause other than run/walk/jog or strenuous sport participation. Subjects not reporting a musculoskeletal injury in the previous 12 months were classified as uninjured. Physical activity Physical activity status was categorized into four mutually exclusive groups according to the usual type of physical activity reported during the preceding 12 months. Sedentary subjects reported no participation in a run/walk/jog program, strenuous sports, racquet sports, bicycling, or swimming. Walkers reported participation in a run/walk/jog program with an average pace per mile of 15 minutes or slower (1 mile = 1.6 km). Runners participated in a run/walk/jog program and reported an average pace of less than 15 minutes/mile. For run/walk/jog participants who did not report information about their pace (4–5 percent of the sample), walking and running status were identified by a question that asked how much time subjects spent walking during their run/walk/jog program. Walkers were those who reported walking for at least 75 percent of their run/walk/jog program, and runners walked less than 25 percent of their run/walk/jog program. Run/walk/jog participants with missing information on pace and who reported walking between 25 and 75 percent of their run/walk/jog program (n = 165) could not be clearly classified as a runner or walker and were excluded from this analysis. Sport participants were those who exclusively engaged in one or more sporting activities (e.g., strenuous sports, racquet sports, bicycling, or swimming) but who did not participate in a run/walk/jog program. The weekly duration of physical activity was calculated by summing the hours of all physical activity per week, including time spent participating in a run/walk/jog program, racquet sports, strenuous sports, swimming, and bicycling. Subjects were classified into physical activity categories based on their total reported hours per week. The low duration group included subjects with a total physical activity time of less than 1.25 hours/week. The moderate duration group was defined as those subjects who reported 1.25–3.75 hours/week, and the high duration group comprised those who reported more than 3.75 hours/week. These categories correspond with less than 15, 15–45, and greater than 45 minutes per day of physical activity for 5 days per week. Cardiorespiratory fitness Cardiorespiratory fitness was assessed at the baseline clinic visit by walking on a motorized treadmill during a physician-supervised graded exercise examination using the modified Balke testing protocol (22). Subjects were monitored for blood pressure, heart rate, and rhythm by a trained technician before and during the test (23). The treadmill test began at a speed of 3.3 miles per hour and 0 percent grade for the first minute. The grade was raised to 2 percent during the second minute, and thereafter the grade increased 1 percent every minute for 25 minutes. At 25 minutes, the grade remained constant and the speed increased 0.2 miles per hour each minute until the subject reached volitional fatigue or was stopped by the physician for medical reasons (23). All subjects reached at least 85 percent of their age-predicted (220 – age) maximum heart rate. The maximal time on the treadmill in seconds was the measure used to classify cardiorespiratory fitness. The total treadmill time is highly correlated with maximal oxygen uptake in men (r = 0.92) and women (r = 0.94) (24, 25). Cardiorespiratory fitness levels were categorized by age and gender-specific quintiles of total treadmill time. The lowest quintile (bottom 20 percent) was classified as low cardiorespiratory fitness. The middle two quintiles (middle 40 percent) and the top two quintiles (top 40 percent) were classified as moderate and high cardiorespiratory fitness, respectively. The low cardiorespiratory fitness group in this cohort has been previously established as a high-risk category for several health outcomes, including mortality and cardiovascular disease, among this same cohort (2, 26, 27). Covariates Age (years) and body weight (pounds; 1 pound = 0.45 kg) were self-reported on the follow-up survey. Height (inches; 1 inch = 0.025 m) was measured at the baseline or during a subsequent clinic visit. Body mass index was calculated as weight (kg)/height (m)2. Body mass index was dichotomized as less than 25 (normal) and 25 or greater (overweight/obese). Other covariates included a history of a previous musculoskeletal injury (prior to 1985) and participation (yes/no) in weight training, stretching, or calisthenic exercises at least twice per week during the preceding 12 months. Statistical analyses Frequency tables were developed for all study variables and are presented by injury type for men and women separately. Polytomous logistic regression was used to assess the relation between injury status (activity related, “other,” and none) and each of the three exposures: the type and duration of physical activity and cardiorespiratory fitness level. The potential modifying effects of body mass index on the relation among physical activity, cardiorespiratory fitness, and injury status were assessed by the inclusion of interaction terms in the regression models using the dichotomous body mass index variable. The following variables were included in the final model as categorical variables: history of previous injury and participation in weight training, stretching, and calisthenic exercises. Age (years), cardiorespiratory fitness level (seconds), and body mass index were included in the final model as continuous variables. Statistical significance was set at the α = 0.05 level for all analyses. All analyses were conducted using SAS statistical software (SAS Institute, Inc., Cary, North Carolina). RESULTS The distribution of descriptive characteristics by injury type for men and women is shown in table 1. Among men, 16 percent reported an activity-related injury and 8 percent reported an “other” injury in the 12 months preceding the survey. Among women, 14 percent reported an activity-related injury and 9 percent reported an “other” injury. As age increased among men and women, the prevalence of activity-related injuries decreased, while the prevalence of “other” injuries was relatively stable among men. Among women aged 41–60 years, the prevalence of “other” injury was more than twice as high as among those aged 20–40 years and nearly twice as high among women over age 60. The proportion of men and women with activity-related injuries increased with increasing duration of physical activity and higher cardiorespiratory fitness levels. The majority of all subjects with either activity-related or “other” injuries reported having a previous injury. TABLE 1. Characteristics of subjects according to injury type and gender, Aerobics Center Longitudinal Study, 1970–1986 Characteristic  Injury type   Men (n = 4,034)   Women (n = 967)   Activity related   Other   None   Activity related   Other   None   No.  %  No.  %  No.  %  No.  %  No.  %  No.  %  Age (years)                           20–40  84  13.0  25  7.4  245  8.0  27  20.3  6  6.9  106  14.2   41–60  460  71.1  233  69.1  1,980  64.9  90  67.7  70  80.5  470  62.9   >60  103  15.9  79  23.4  825  27.0  16  12.0  11  12.6  171  22.9  Body mass index*                           Normal  346  53.5  143  42.4  1,388  45.5  110  82.7  66  75.9  620  83.0   Overweight/obese  301  46.5  194  57.6  1,662  54.5  23  17.3  21  24.1  127  17.0  PA† type                           Sedentary  111  17.2  132  39.2  1,172  38.4  36  27.1  41  47.1  327  43.8   Walker  44  6.8  32  9.5  307  10.1  27  20.3  13  14.9  164  22.0   Runner  416  64.3  135  40.1  1,305  42.8  59  44.4  25  28.7  210  28.1   Sport  76  11.7  39  11.6  266  8.7  11  8.3  8  9.2  46  6.2  PA duration                           Low  186  28.7  170  50.4  1,588  52.1  46  34.6  50  57.5  395  52.9   Moderate  227  35.1  98  29.1  932  30.6  40  30.1  21  24.1  218  29.2   High  234  36.2  69  20.5  530  17.4  47  35.3  16  18.4  134  17.9  Fitness level                           Low  24  3.7  56  16.6  422  13.8  11  8.3  16  18.4  110  14.7   Moderate  153  23.6  103  30.6  1,089  35.7  43  32.3  33  37.9  302  40.4   High  470  72.6  178  52.8  1,539  50.5  79  59.4  38  43.7  335  44.8  Previous injury                           Yes  503  77.8  236  70.0  1,691  55.4  86  64.7  54  62.1  283  37.9   No  144  22.3  101  30.0  1,359  44.6  47  35.3  33  37.9  464  62.1  Weight training                           Yes  396  61.2  268  79.5  2,397  78.6  95  71.4  76  87.4  664  88.9   No  251  38.8  69  20.5  653  20.8  38  28.6  11  12.6  83  11.1  Stretching                           Yes  263  40.6  202  59.9  1,942  63.7  58  43.6  55  63.2  462  61.8   No  384  59.4  135  40.1  1,108  36.3  75  56.4  32  36.8  285  3.7  Calisthenics                           Yes  471  72.8  283  84.0  2,439  80.0  88  66.2  70  80.5  585  78.3   No  176  27.2  54  16.0  611  20.0  45  33.8  17  19.5  162  21.7   Total  647    337    3,050    133    87    747    Characteristic  Injury type   Men (n = 4,034)   Women (n = 967)   Activity related   Other   None   Activity related   Other   None   No.  %  No.  %  No.  %  No.  %  No.  %  No.  %  Age (years)                           20–40  84  13.0  25  7.4  245  8.0  27  20.3  6  6.9  106  14.2   41–60  460  71.1  233  69.1  1,980  64.9  90  67.7  70  80.5  470  62.9   >60  103  15.9  79  23.4  825  27.0  16  12.0  11  12.6  171  22.9  Body mass index*                           Normal  346  53.5  143  42.4  1,388  45.5  110  82.7  66  75.9  620  83.0   Overweight/obese  301  46.5  194  57.6  1,662  54.5  23  17.3  21  24.1  127  17.0  PA† type                           Sedentary  111  17.2  132  39.2  1,172  38.4  36  27.1  41  47.1  327  43.8   Walker  44  6.8  32  9.5  307  10.1  27  20.3  13  14.9  164  22.0   Runner  416  64.3  135  40.1  1,305  42.8  59  44.4  25  28.7  210  28.1   Sport  76  11.7  39  11.6  266  8.7  11  8.3  8  9.2  46  6.2  PA duration                           Low  186  28.7  170  50.4  1,588  52.1  46  34.6  50  57.5  395  52.9   Moderate  227  35.1  98  29.1  932  30.6  40  30.1  21  24.1  218  29.2   High  234  36.2  69  20.5  530  17.4  47  35.3  16  18.4  134  17.9  Fitness level                           Low  24  3.7  56  16.6  422  13.8  11  8.3  16  18.4  110  14.7   Moderate  153  23.6  103  30.6  1,089  35.7  43  32.3  33  37.9  302  40.4   High  470  72.6  178  52.8  1,539  50.5  79  59.4  38  43.7  335  44.8  Previous injury                           Yes  503  77.8  236  70.0  1,691  55.4  86  64.7  54  62.1  283  37.9   No  144  22.3  101  30.0  1,359  44.6  47  35.3  33  37.9  464  62.1  Weight training                           Yes  396  61.2  268  79.5  2,397  78.6  95  71.4  76  87.4  664  88.9   No  251  38.8  69  20.5  653  20.8  38  28.6  11  12.6  83  11.1  Stretching                           Yes  263  40.6  202  59.9  1,942  63.7  58  43.6  55  63.2  462  61.8   No  384  59.4  135  40.1  1,108  36.3  75  56.4  32  36.8  285  3.7  Calisthenics                           Yes  471  72.8  283  84.0  2,439  80.0  88  66.2  70  80.5  585  78.3   No  176  27.2  54  16.0  611  20.0  45  33.8  17  19.5  162  21.7   Total  647    337    3,050    133    87    747    * Body mass index categories are defined as normal (<25 kg/m2) and overweight/obese (≥25 kg/m2). † PA, physical activity. View Large The distribution of activity-related and “other” injuries by body site among all subjects is presented in figure 1. Activity-related injuries were more common in the foot, ankle, lower leg, and thigh than “other” injuries. The proportion of “other” back and arm injuries was higher than the proportion of the activity-related back and arm injuries. The frequency of knee, hip, wrist, and hand injuries was similar for activity-related and “other” injuries. FIGURE 1. View largeDownload slide Distribution of activity-related and “other” injuries by body site among men and women, Aerobics Center Longitudinal Study, 1970–1986. FIGURE 1. View largeDownload slide Distribution of activity-related and “other” injuries by body site among men and women, Aerobics Center Longitudinal Study, 1970–1986. The independent associations between physical activity type, duration, and cardiorespiratory fitness level and musculoskeletal injury in men and women are shown in table 2. Body mass index did not modify the association between physical activity or cardiorespiratory fitness level and injury status and, therefore, regression models were adjusted for age, previous injury, body mass index, weight training, stretching, and calisthenic exercises. Running and sport participation were associated with approximately 2.3 times higher risk of activity-related injury in men and with 1.5–1.7 times higher risk in women. No significant difference was found between walkers and sedentary subjects for the risk of activity-related injury in women; male walkers had 73 percent higher risk of activity-related injury than did sedentary men. TABLE 2. Odds ratios and 95 percent confidence intervals for activity-related and other injury and for type and duration of physical activity and cardiorespiratory fitness among 5,001 men and women, Aerobics Center Longitudinal Study, 1970–1986   No.  Activity-related injury   Other injury     OR*,†  95% CI*  OR†  95% CI  Men             PA* type              Sedentary  1,415  1.00    1.00      Walker  383  1.73  1.18, 2.54  1.01  0.67, 1.54    Runner  1,856  2.38  1.87, 3.04  0.87  0.68, 1.11    Sport  381  2.26  1.62, 3.15  1.20  0.81, 1.77   PA duration              Low  1,944  1.00    1.00      Moderate  1,257  1.71  1.37, 2.13  0.98  0.74, 1.28    High  833  2.78  2.21, 3.50  1.18  0.86, 1.60   Fitness level              Low  502  1.00    1.00      Moderate  1,345  2.35  1.50, 3.70  0.70  0.50, 1.00    High  2,187  3.99  2.58, 6.18  0.84  0.60, 1.17  Women             PA type              Sedentary  404  1.00    1.00      Walker  204  1.35  0.77, 2.35  0.65  0.34, 1.27    Runner  294  1.68  1.03, 2.74  0.96  0.54, 1.70    Sport  65  1.50  0.68, 3.28  1.38  0.59, 3.25   PA duration              Low  491  1.00    1.00      Moderate  279  1.33  0.82, 2.14  0.79  0.45, 1.37    High  197  2.03  1.25, 3.30  0.87  0.46, 1.62   Fitness level              Low  137  1.00    1.00      Moderate  378  1.26  0.61, 2.58  0.79  0.41, 1.52    High  452  1.78  0.88, 3.58  0.85  0.44, 1.64    No.  Activity-related injury   Other injury     OR*,†  95% CI*  OR†  95% CI  Men             PA* type              Sedentary  1,415  1.00    1.00      Walker  383  1.73  1.18, 2.54  1.01  0.67, 1.54    Runner  1,856  2.38  1.87, 3.04  0.87  0.68, 1.11    Sport  381  2.26  1.62, 3.15  1.20  0.81, 1.77   PA duration              Low  1,944  1.00    1.00      Moderate  1,257  1.71  1.37, 2.13  0.98  0.74, 1.28    High  833  2.78  2.21, 3.50  1.18  0.86, 1.60   Fitness level              Low  502  1.00    1.00      Moderate  1,345  2.35  1.50, 3.70  0.70  0.50, 1.00    High  2,187  3.99  2.58, 6.18  0.84  0.60, 1.17  Women             PA type              Sedentary  404  1.00    1.00      Walker  204  1.35  0.77, 2.35  0.65  0.34, 1.27    Runner  294  1.68  1.03, 2.74  0.96  0.54, 1.70    Sport  65  1.50  0.68, 3.28  1.38  0.59, 3.25   PA duration              Low  491  1.00    1.00      Moderate  279  1.33  0.82, 2.14  0.79  0.45, 1.37    High  197  2.03  1.25, 3.30  0.87  0.46, 1.62   Fitness level              Low  137  1.00    1.00      Moderate  378  1.26  0.61, 2.58  0.79  0.41, 1.52    High  452  1.78  0.88, 3.58  0.85  0.44, 1.64  * OR, odds ratio; CI, confidence interval; PA, physical activity. † Each type, duration, and fitness models are separate models and each is adjusted for age, body mass index, previous injury, weight training, stretching, and calisthenic exercises. View Large In men with activity-related injuries, both moderate and high duration of physical activity per week and moderate and high cardiorespiratory fitness levels were associated with a significant increased risk of musculoskeletal injury compared with subjects in the low categories of duration and cardiorespiratory fitness. The risk of activity-related injury among men in the highest duration category was 2.8 times as high as for men in the low duration category. Highly fit men had the greatest risk of musculoskeletal injury, almost four times that of men in the lowest cardiorespiratory fitness category. Men with a moderate duration of physical activity per week (1.25–3.75 hours/week) and moderate cardiorespiratory fitness levels had about half the risk of activity-related injury as did those in the high duration and cardiorespiratory fitness categories. In women, the high level of duration and cardiorespiratory fitness was significantly associated with an increased risk of activity-related injury compared with women who had a low level of duration and cardiorespiratory fitness. The multivariate relation between activity-related injury and activity type, duration, and cardiorespiratory fitness in nonsedentary subjects (n = 2,931) is shown in table 3. Compared with that in walkers, the risk of activity-related injuries was not different for runners and sport participants for any duration of activity per week. Among walkers and sport participants, an increasing duration of activity per week did not significantly increase the risk of injury. In contrast, among runners the risk of injury increased with increasing duration of activity per week. TABLE 3. Odds ratios and 95 percent confidence intervals for the relation between activity-related injury and type of physical activity stratified by duration of physical activity and the relation between activity-related injury and duration of physical activity stratified by type of physical activity among nonsedentary subjects, Aerobics Center Longitudinal Study, 1970–1986   No.  Duration of PA*     <1.25 hours/week   1.25–3.75 hours/week   >3.75 hours/week     OR*,†  95% CI*  OR†  95% CI  OR†  95% CI  Type of PA                 Walker  542  1.00    1.00    1.00     Runner  1,990  0.45  0.20, 1.01  1.38  0.84, 2.27  1.40  0.87, 2.26   Sport  399   0.66   0.17, 2.48   1.15   0.62, 2.13   1.09   0.65, 1.83     No.  Type of PA     Walker   Runner   Sport     OR†   95% CI*   OR†   95% CI   OR†   95% CI   Duration of PA                 <1.25 hours/week  569  1.00    1.00    1.00     1.25–3.75 hours/week  1,417  0.66  0.31, 1.43  1.47  1.08, 2.00  0.91  0.27, 3.05   >3.75 hours/week  945  1.16  0.54, 2.52  2.38  1.72, 3.29  1.30  0.40, 4.23    No.  Duration of PA*     <1.25 hours/week   1.25–3.75 hours/week   >3.75 hours/week     OR*,†  95% CI*  OR†  95% CI  OR†  95% CI  Type of PA                 Walker  542  1.00    1.00    1.00     Runner  1,990  0.45  0.20, 1.01  1.38  0.84, 2.27  1.40  0.87, 2.26   Sport  399   0.66   0.17, 2.48   1.15   0.62, 2.13   1.09   0.65, 1.83     No.  Type of PA     Walker   Runner   Sport     OR†   95% CI*   OR†   95% CI   OR†   95% CI   Duration of PA                 <1.25 hours/week  569  1.00    1.00    1.00     1.25–3.75 hours/week  1,417  0.66  0.31, 1.43  1.47  1.08, 2.00  0.91  0.27, 3.05   >3.75 hours/week  945  1.16  0.54, 2.52  2.38  1.72, 3.29  1.30  0.40, 4.23  * PA, physical activity; OR, odds ratio; CI, confidence interval. † Adjusted for cardiorespiratory fitness, body mass index, age, previous injury, weight training, stretching, and calisthenic exercises. View Large DISCUSSION This is the first report of the risk of musculoskeletal injury associated with selected types and duration of physical activity and cardiorespiratory fitness among recreationally active adults. The risk of activity-related injury among adults increased for runners, sport participants, persons engaging in more than 1.25 hours per week of physical activity, and persons with moderate to high cardiorespiratory fitness levels. Walking as a primary mode of physical activity was not associated with a significant increase in risk of injury among women, and injury risk did not increase with longer durations of walking per week. The foot and ankle were the most frequent sites of activity-related injuries while the back was the most common site of “other” injury. Musculoskeletal injuries among physically active populations tend to occur in the lower extremity (7, 9–19). Activity-related back injuries, however, tend to be sport specific, because sports such as gymnastics, golf, and weight lifting have a higher incidence of back injuries than other sports (28, 29). Population-based and health club studies describing the distribution of general musculoskeletal injuries among adults also report the back as one of the top three sites of injury (30, 31). In this study we measured four of five components of physical activity that can contribute to the risk of musculoskeletal injury: cardiorespiratory fitness and type, frequency, and duration of activity. We do not have a measure of the intensity of exercise for all subjects, primarily those reporting participation in sports activities such as swimming, cycling, and other strenuous sports. Our findings regarding high cardiorespiratory fitness level as a risk factor for injury contrast with results from military studies reporting low cardiorespiratory fitness levels being associated with increased training injuries (13, 20, 21). One explanation may be that the cardiorespiratory fitness level in our study is acting as a surrogate for activity intensity or some other unmeasured component of activity. However, among nonsedentary subjects, a higher cardiorespiratory fitness level was still associated with an increased risk of activity-related injury, even after adjusting for the type and duration of physical activity. Furthermore, among just the runners and walkers, cardiorespiratory fitness was still associated with a nonsignificant 50–74 percent increased risk of injury after controlling for type (running vs. walking), duration (hours per week of activity), and intensity (average pace). Even with these findings, these results suggest that other factors, such as biomechanical or structural abnormalities or personality traits, may also play a role in the relation between physical activity and the risk of musculoskeletal injuries. Differences in the study design and sample between our study and military studies may also explain our contrasting findings regarding cardiorespiratory fitness as a risk factor for injury. The military population is young (18–23 years old), while our subjects range from 20 to 85 years. Being older on average, our subjects may have sustained more injuries over their lifetime than younger persons, thereby increasing their risk of subsequent injuries. We controlled for previous injuries. However, some subjects may not have reported injuries that occurred a long time in the past. In military studies, cardiorespiratory fitness was measured immediately preceding entrance into a vigorous basic training program, and subjects most likely did not have sufficient exercise experience prior to testing to have high cardiorespiratory fitness levels. Because their lifetime exercise experience is higher, our subjects' baseline cardiorespiratory fitness levels may have been higher than those of military trainees. Cardiorespiratory fitness is usually measured by 2-mile-run times in military studies, and therefore we cannot directly compare our baseline cardiorespiratory fitness measure (seconds on treadmill) with theirs. It would be interesting to know how cardiorespiratory fitness measured after a 6-week basic military training program, when subjects have accommodated to the increased activity levels, correlates with injuries reported over the next 12 months. When measured over a year as in our study, it may be that even military trainees with high cardiorespiratory fitness levels will be at high risk of injury. Military trainees engage in a well-controlled physical training program, and training volume is essentially equal among subjects. Our subjects are free to vary their activity level and therefore have a wide range of activity time per week. One military study that did report some variation in hours per week of training found, similar to ours, that training time per week was correlated with injuries (32). In addition, military studies use a stricter definition of injury requiring that an injury must have required medical attention and be recorded in the medical record. We included any injury the subject attributed to physical activity. In summary, the association of a higher level of cardiorespiratory fitness with higher injury rates in this cohort most likely reflects our subjects' high lifetime activity levels. A higher level of cardiorespiratory fitness may also be acting as a surrogate for some unmeasured risk factor for injury such as biomechanical or structural abnormalities, personality traits, or exercise behavior patterns. Identifying activities that provide an adequate amount of physical activity to improve health but have a lower risk of injury is an important public health goal. Walking is a popular form of moderate physical activity; an estimated 138 million US adults report walking for exercise (33). According to recent national estimates, 1.4 percent of US adults are injured while walking for exercise (33). Several exercise training studies report a lower frequency of injury during walking activities than during jogging (34–36). In an epidemiologic study comparing runners and walkers, walkers had a lower risk of activity-related injury (37). In addition, the injury risk among runners significantly increased with increasing duration of running per day, while increased duration of walking per day did not significantly increase the injury risk among walkers. We report similar findings that, among walkers and sport participants, the risk of injury did not increase with increased duration of activity per week. In contrast, among runners the risk of activity-related injury increases with longer duration of physical activity per week in a classic dose-response pattern. None of the relations among physical activity, cardiorespiratory fitness level, and the risk of “other” injuries was statistically significant, which is likely due to the small number of “other” injuries reported (approximately 8 percent). However, a consistent trend is evident, suggesting that the type of physical activity and cardiorespiratory fitness level may provide some protection against an “other” injury. Running decreased the risk of an “other” injury 13 percent among men, and walking decreased the risk of “other” injury 35 percent among women. Moderate cardiorespiratory fitness also decreased the risk of an “other” injury 30 percent among men and 21 percent among women. A strong and efficient neuromuscular system is better able to adapt to acute stresses and subsequently may be more resistant to physical injury (38–41). Thus, moderate physical activity in amounts to achieve a moderate level of cardiorespiratory fitness may provide an adequate stimulus to strengthen the neuromuscular and skeletal systems and decrease the risk of injury, regardless of the cause of injury. In addition to the small number of “other” injuries reported, several other limitations of these data should be mentioned. First, inadequate recall of information used to classify each subject into exposure and outcome groups may introduce measurement error (42). Measurement error in this sample is expected to be low because the accuracy of recall of self-reported health information among subjects in the Aerobics Center Longitudinal Study has been previously reported as excellent (98 percent sensitivity, 99 percent specificity) (1). Second, subjects reported if an injury was related to participation in a run/walk/jog program or a strenuous sport program or due to some “other” cause. The cause of injuries reported in the “other” category was not specified. Therefore, some injuries due to sports activities, such as golf, may not have been perceived by subjects to fit into the run/walk/jog or strenuous sports categories. These other types of sports injuries may have gone unreported or may have been included in the “other” category and, therefore, it is possible that some activity-related injuries may have been classified as other injuries. This type of misclassification would most likely be nondifferential and would bias the risk estimate toward the null. Third, 8 percent of sedentary subjects reported an activity-related injury. The sedentary category included subjects who did not report any participation in the specific activities listed on the survey. Therefore, sedentary subjects did not consider themselves to be regularly physically active but may have been injured during occasional participation in sport activities on weekends or at social gatherings. Lastly, cardiorespiratory fitness levels may change significantly from baseline to follow-up, an average of 8.2 years in this sample. However, a unique feature of this cohort is that cardiorespiratory fitness levels have been reported to be relatively constant over time. Using subjects from the same cohort, Blair et al. (43) report that approximately 82 percent of subjects remained fit from the baseline treadmill examination to a second treadmill examination (the mean time between examinations was 5 years). In our sample, using body mass index as a proxy for cardiorespiratory fitness, over 90 percent of subjects maintained stable weight from baseline to follow-up. An additional limitation is that, although the response rate was adequate, differences may exist between the subjects who responded to the follow-up survey and those who did not respond. Macera et al. (44) conducted a comparison of subjects who did and did not return the 1982 follow-up survey in this same cohort. Using baseline laboratory measurements and a personal health history, nonrespondents and respondents were equally healthy. However, participants reporting more positive health behaviors and more family history of chronic disease were more likely to respond to the mail survey. Assuming that the response patterns reported by Macera et al. are similar to those of the current study using the 1986 follow-up survey, subjects included in this sample most likely represent health conscious adults who demonstrate behaviors consistent with good health, such as engaging in physical activity and not smoking. Thus, the results of this study may not be applicable to all segments of the population. Several aspects of these data and the statistical procedures contribute to the strengths of this study. Detailed data on physical activity, cardiorespiratory fitness, and injury were obtained from a large sample of adult men and women. Subjects reported a wide range of activity levels, including no regular physical activity, which allowed comparison of injury experiences across different levels and duration of activity. Cardiorespiratory fitness was directly measured by a treadmill graded exercise test and serves as an objective measurement of cardiorespiratory fitness level. Simultaneously estimating the association between each exposure and activity-related and “other” injuries using polytomous logistic regression allowed for direct comparison of the risk estimates for each injury cause (45). It is well known that subgroups of vigorously active populations such as runners, scholastic athletes, and military trainees experience high rates of musculoskeletal injury (7, 9–19, 21). The results of this study indicate that the moderate types and duration of physical activity promoted by private and federal health organizations have lower injury risk than more vigorous types and longer durations of physical activity. Walking for exercise was not associated with a significant increased risk of activity-related injury, even among those walkers with the highest duration of activity per week. This suggests that for most adults, walking is a safe form of physical activity. However, to minimize the risk of injury, public health practitioners need to target injury prevention messages to adults engaged in vigorous types of physical activity. Reprint requests to Dr. Jennifer M. Hootman, Centers for Disease Control and Prevention, 4770 Buford Highway N.E., MS K-45, Atlanta, GA 30341 (e-mail: [email protected]). This study was supported by grant NIH NIA AG06945 from the National Institutes of Health awarded to the Cooper Institute for Aerobics Research. The authors thank Dorothy Davis of the University of South Carolina Prevention Research Center and the staff of the Cooper Institute for Aerobics Research for collecting and managing the data used in this study. REFERENCES 1. Blair S, Goodyear N, Gibbons L, et al. Physical fitness and incidence of hypertension in healthy normotensive men and women. JAMA  1984; 252: 487–90. Google Scholar 2. Blair S, Kohl H, Paffenbarger R, et al. Physical fitness and all-cause mortality: a prospective study of healthy men and women. JAMA  1989; 262: 2395–401. Google Scholar 3. US Department of Health and Human Services. Physical activity and health: a report of the Surgeon General. Atlanta, GA: Centers for Disease Control and Prevention, 1996. Google Scholar 4. American College of Sports Medicine. The recommended quantity and quality of exercise for developing and maintaining cardiorespiratory and muscular fitness and flexibility in healthy adults. Med Sci Sports Exerc  1998; 30: 975–91. Google Scholar 5. Pate R, Pratt M, Blair S, et al. Physical activity and public health: a recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine. JAMA  1995; 273: 402–7. Google Scholar 6. Uitenbroek D. Sports, exercise, and other causes of injuries: results of a population study. Res Q Exerc Sport  1996; 67: 380–5. Google Scholar 7. Koplan J, Powell K, Sikes R, et al. An epidemiologic study of the benefits and risks of running. JAMA  1982; 248: 3118–21. Google Scholar 8. Koplan J, Siscovick D, Goldbaum G. The risks of exercise: a public health view of injuries and hazards. Public Health Rep  1985; 100: 189–95. Google Scholar 9. Bovens A, Janssen G, Vermeer H, et al. Occurence of running injuries in adults following a supervised training program. Int J Sports Med  1989; 10(suppl): S186–90. Google Scholar 10. Caselli M, Longobardi S. Lower extremity injuries at the New York City marathon. J Am Podiatr Med Assoc  1997; 87: 34–7. Google Scholar 11. Blair S, Kohl H, Goodyear N. Rates and risks for running and exercise injuries: studies in three populations. Res Q Exerc Sport  1987; 58: 221–8. Google Scholar 12. Jacobs S, Berson B. Injuries to runners: a study of entrants to a 10,000 meter race. Am J Sports Med  1986; 14: 151–5. Google Scholar 13. Jones BH, Bovee MW, Harris J, et al. Intrinsic risk factors for exercise-related injuries among male and female Army trainees. Am J Sports Med  1993; 21: 705–10. Google Scholar 14. Lysholm J, Wiklander J. Injuries in runners. Am J Sports Med  1987; 15: 168–71. Google Scholar 15. Macera C, Pate R, Powell K, et al. Predicting lower-extremity injuries among habitual runners. Arch Intern Med  1989; 149: 2565–8. Google Scholar 16. Macera C, Pate R, Woods J, et al. Postrace morbidity among runners. Am J Prev Med  1991; 7: 194–8. Google Scholar 17. Marti B, Vader J, Minder C, et al. On the epidemiology of running injuries: the 1984 Bern Grand-Prix Study. Am J Sports Med  1988; 16: 285–94. Google Scholar 18. Marti B. Benefits and risks of running among women: an epidemiologic study. Int J Sports Med  1988; 9: 92–8. Google Scholar 19. Walter S, Hart L, McIntosh J. The Ontario cohort study running-related injuries. Arch Intern Med  1989; 149: 2561–4. Google Scholar 20. Shaffer RA, Brodine SK, Almeida SA, et al. Use of measures of physical activity to predict stress fractures in young men undergoing a rigorous physical training program. Am J Epidemiol  1999; 149: 236–42. Google Scholar 21. Knapik J, Ang P, Reynolds K, et al. Physical fitness, age, and injury incidence in infantry soldiers. J Med  1993; 35: 598–603. Google Scholar 22. Balke B, Ware R. An experimental study of physical fitness on Air Force personnel. US Armed Forces Med J  1959; 10: 675–88. Google Scholar 23. Cooper Institute for Aerobics Research. Procedures manual for Cooper Clinic technicians. Dallas, TX: Cooper Clinic, 1986. Google Scholar 24. Pollock M, Bohannon R, Cooper K, et al. A comparative analysis of four protocols for maximal treadmill stress testing. Am Heart J  1976; 92: 39–46. Google Scholar 25. Pollock M, Foster C, Schmidt D, et al. Comparative analysis of physiologic responses to three different maximal graded exercise test protocols in healthy women. Am Heart J  1982; 103: 363–73. Google Scholar 26. Lee CD, Blair SN, Jackson AS. Cardiorespiratory fitness, body composition, and all-cause and cardiovascular disease mortality in men. Am J Clin Nutr  1999; 69: 373–80. Google Scholar 27. Blair S, Kohl H, Barlow C, et al. Changes in physical fitness and all-cause mortality: a prospective study of healthy and unhealthy men. JAMA  1995; 273: 1093–8. Google Scholar 28. Caine D, Caine C, Linder K. The epidemiology of sports injuries. Champaign, IL: Human Kinetics Press, 1996. Google Scholar 29. Wadley GH, Albright JP. Women's intercollegiate gymnastics: injury patterns and “permanent” medical disability. Am J Sports Med  1993; 21: 314–20. Google Scholar 30. vanMechelen W, Twisk J, Molendijk A, et al. Subject-related risk factors for sports injuries: a 1-year prospective study in young adults. Med Sci Sports Exerc  1996; 28: 1171–9. Google Scholar 31. Sedgwick A, Smith D, Davies M. Musculoskeletal status of men and women who entered a fitness program. Med J Aust  1988; 148: 385–91. Google Scholar 32. Almeida SA, Williams, KM, Shaffer RA, et al. Epidemiological patterns of musculoskeletal injuries and physical training. Med Sci Sports Exerc  1999; 31: 1176–82. Google Scholar 33. Powell K, Heath G, Kresnow M, et al. Injury rates from walking, gardening, weightlifting, outdoor bicycling, and aerobics. Med Sci Sports Exerc  1998; 30: 1246–9. Google Scholar 34. Pollock M, Carroll J, Graves J, et al. Injuries and adherence to walk/jog and resistance training programs in the elderly. Med Sci Sports Exerc  1991; 23: 1194–200. Google Scholar 35. Suter E, Marti B, Gutzwiller F. Jogging or walking—comparison of health effects. Ann Epidemiol  1994; 4: 375–81. Google Scholar 36. Carroll J, Pollock M, Graves J, et al. Incidence of injury during moderate and high-intensity walking training in the elderly. J Gerontol  1992; 47: M61–6. Google Scholar 37. Colbert LH, Hootman JM, Macera CA. Physical activity-related injuries in walkers and runners in the Aerobics Center Longitudinal Study. Clin J Sport Med  2000; 10: 259–63. Google Scholar 38. Radin E. Role of muscles in protecting athletes from injury. Acta Med Scand  1986; 711: 143–7. Google Scholar 39. Sale D. Neural adaptation to resistance training. Med Sci Sports Exerc  1988; 20(suppl): S135–45. Google Scholar 40. Vuori I. Exercise and physical health: musculoskeletal health and functional capabilities. Res Q Exerc Sport  1995; 66: 276–85. Google Scholar 41. Wolfson L, Judge J, Whipple R, et al. Strength is a major factor in balance, gait, and the occurrence of falls. J Gerontol  1995; 50A: 64–7. Google Scholar 42. Gordis L. Epidemiology. Philadelphia, PA: WB Saunders Company, 1996. Google Scholar 43. Blair SN, Kohl HW, Barlow CE, et al. Changes in physical fitness and all-cause mortality: a prospective study of healthy and unhealthy men. JAMA  1995; 273: 1093–8. Google Scholar 44. Macera CA, Jackson K, Davis D, et al. Patterns of non-response to a mail survey. J Clin Epidemiol  1990; 43: 1427–30. Google Scholar 45. Hosmer D, Lemeshow S. Applied logistic regression. New York, NY: John Wiley & Sons, Inc, 1989. Google Scholar http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png American Journal of Epidemiology Oxford University Press

Association among Physical Activity Level, Cardiorespiratory Fitness, and Risk of Musculoskeletal Injury

Download PDF
8 pages

Loading next page...
 
/lp/oxford-university-press/association-among-physical-activity-level-cardiorespiratory-fitness-fxQNX3Y10e

References (43)

Publisher
Oxford University Press
ISSN
0002-9262
eISSN
1476-6256
DOI
10.1093/aje/154.3.251
Publisher site
See Article on Publisher Site

Abstract

Abstract To help public health practitioners promote physical activities with a low risk of injury, this study determined the relation among type and duration of physical activity, cardiorespiratory fitness, and musculoskeletal injury in a sample of adults enrolled in the Aerobics Center Longitudinal Study. Subjects included 4,034 men and 967 women who underwent a baseline physical examination between 1970 and 1985 and who returned a mailed follow-up survey in 1986. At baseline, a treadmill graded exercise test was used to measure cardiorespiratory fitness. At follow-up, subjects reported injuries and type and duration of physical activity in the preceding 12 months. Polytomous logistic regression was used to estimate the association among physical activity type and duration, cardiorespiratory fitness, and injury. The risk of sustaining an activity-related injury increased with higher duration of physical activity per week and cardiorespiratory fitness levels. Results suggest that cardiorespiratory fitness may be a surrogate for unmeasured components of physical activity, such as exercise intensity. Among walkers, increasing duration of activity per week was not associated with an increased risk of injury. Results suggest that, for most adults, walking is a safe form of physical activity associated with a lower risk of injury than running or sport participation. exercise, logistic models, musculoskeletal system, physical fitness, wounds and injuries Physical inactivity is a leading public health problem associated with decreased longevity as well as cardiovascular disease, cancer, obesity, diabetes, and other diseases (1–3). Recently, private and federal organizations have published physical activity recommendations to increase physical activity levels in the US general adult population (3–5). In a joint statement in 1995, the Centers for Disease Control and Prevention and the American College of Sports Medicine recommended that all adults accumulate bouts of moderate physical activity totaling at least 30 minutes on most, if not all, days of the week (5). Walking, jogging, tennis, volleyball, bicycling, and swimming are popular activities that meet these recommendations (3, 5). The risk of injury is one of the few adverse effects of a physically active lifestyle (3, 5–8). Among vigorously active populations, a clear dose-response relation exists between weight-bearing exercise volume and risk of lower extremity injury (7, 9–19). Studies of runners and military trainees have found that higher levels of cardiorespiratory fitness are associated with lower risk of exercise-related musculoskeletal injury (13, 15, 20, 21). However, the relation among physical activity, cardiorespiratory fitness, and musculoskeletal injury has not been defined for the civilian general population that participates in the moderate types of exercise recommended by national health organizations, such as walking, jogging, or sport activities. Identifying the relation between different levels of physical activity and musculoskeletal injury may help public health practitioners promote safe physical activities for adults. The purpose of this study was to identify the association among physical activity type and duration, cardiorespiratory fitness, and musculoskeletal injury in a group of adults with various physical activity and cardiorespiratory fitness levels. MATERIALS AND METHODS Subjects The data for this study were obtained from the Aerobics Center Longitudinal Study, a prospective epidemiologic study investigating the influence of physical activity and cardiorespiratory fitness on health outcomes among healthy adults. Potential subjects attended the Cooper Clinic, a preventive medicine clinic in Dallas, Texas, where they underwent a baseline physical examination, between 1970 and 1981, that included a personal and family health history questionnaire, a physician examination, anthropometric measurements, blood analyses, and a treadmill graded exercise test. Subjects were enrolled in the Aerobics Center Longitudinal Study after reading and signing a consent form. The consent form and the Aerobics Center Longitudinal Study protocol are reviewed annually by the Cooper Institute for Aerobics Research review board. Upon enrollment, the health status of subjects was monitored approximately every 5 years by mailed follow-up surveys. Of the 18,806 baseline clinical examinations through 1981, only subjects (n = 11,972) that returned a follow-up survey in 1982 were mailed the 1986 survey. A total of 5,028 men and 1,283 women returned the 1986 survey. After deceased subjects and incorrect addresses were taken into account, the response rate of the 1986 survey was 72 percent. Subjects eligible for this study were adults aged 20–85 years (at follow-up) who completed the baseline examination and returned a follow-up survey in 1986. Analyses were limited to the 4,034 men and 967 women who returned the survey and had complete data for all study variables. Additional details of the Aerobics Center Longitudinal Study methods have been described previously (1, 2). Injury definition An injury was any self-reported muscle, tendon, bone, ligament, or joint injury occurring in the 12 months before the mailed follow-up survey. The cause of injury was defined as “activity related” if injury was due to participation in a run/walk/jog or strenuous sport exercise program (e.g., racquet sports, cycling, swimming, aerobic dance, basketball, soccer, and so on). It was defined as “other injury” if the injury was due to a cause other than run/walk/jog or strenuous sport participation. Subjects not reporting a musculoskeletal injury in the previous 12 months were classified as uninjured. Physical activity Physical activity status was categorized into four mutually exclusive groups according to the usual type of physical activity reported during the preceding 12 months. Sedentary subjects reported no participation in a run/walk/jog program, strenuous sports, racquet sports, bicycling, or swimming. Walkers reported participation in a run/walk/jog program with an average pace per mile of 15 minutes or slower (1 mile = 1.6 km). Runners participated in a run/walk/jog program and reported an average pace of less than 15 minutes/mile. For run/walk/jog participants who did not report information about their pace (4–5 percent of the sample), walking and running status were identified by a question that asked how much time subjects spent walking during their run/walk/jog program. Walkers were those who reported walking for at least 75 percent of their run/walk/jog program, and runners walked less than 25 percent of their run/walk/jog program. Run/walk/jog participants with missing information on pace and who reported walking between 25 and 75 percent of their run/walk/jog program (n = 165) could not be clearly classified as a runner or walker and were excluded from this analysis. Sport participants were those who exclusively engaged in one or more sporting activities (e.g., strenuous sports, racquet sports, bicycling, or swimming) but who did not participate in a run/walk/jog program. The weekly duration of physical activity was calculated by summing the hours of all physical activity per week, including time spent participating in a run/walk/jog program, racquet sports, strenuous sports, swimming, and bicycling. Subjects were classified into physical activity categories based on their total reported hours per week. The low duration group included subjects with a total physical activity time of less than 1.25 hours/week. The moderate duration group was defined as those subjects who reported 1.25–3.75 hours/week, and the high duration group comprised those who reported more than 3.75 hours/week. These categories correspond with less than 15, 15–45, and greater than 45 minutes per day of physical activity for 5 days per week. Cardiorespiratory fitness Cardiorespiratory fitness was assessed at the baseline clinic visit by walking on a motorized treadmill during a physician-supervised graded exercise examination using the modified Balke testing protocol (22). Subjects were monitored for blood pressure, heart rate, and rhythm by a trained technician before and during the test (23). The treadmill test began at a speed of 3.3 miles per hour and 0 percent grade for the first minute. The grade was raised to 2 percent during the second minute, and thereafter the grade increased 1 percent every minute for 25 minutes. At 25 minutes, the grade remained constant and the speed increased 0.2 miles per hour each minute until the subject reached volitional fatigue or was stopped by the physician for medical reasons (23). All subjects reached at least 85 percent of their age-predicted (220 – age) maximum heart rate. The maximal time on the treadmill in seconds was the measure used to classify cardiorespiratory fitness. The total treadmill time is highly correlated with maximal oxygen uptake in men (r = 0.92) and women (r = 0.94) (24, 25). Cardiorespiratory fitness levels were categorized by age and gender-specific quintiles of total treadmill time. The lowest quintile (bottom 20 percent) was classified as low cardiorespiratory fitness. The middle two quintiles (middle 40 percent) and the top two quintiles (top 40 percent) were classified as moderate and high cardiorespiratory fitness, respectively. The low cardiorespiratory fitness group in this cohort has been previously established as a high-risk category for several health outcomes, including mortality and cardiovascular disease, among this same cohort (2, 26, 27). Covariates Age (years) and body weight (pounds; 1 pound = 0.45 kg) were self-reported on the follow-up survey. Height (inches; 1 inch = 0.025 m) was measured at the baseline or during a subsequent clinic visit. Body mass index was calculated as weight (kg)/height (m)2. Body mass index was dichotomized as less than 25 (normal) and 25 or greater (overweight/obese). Other covariates included a history of a previous musculoskeletal injury (prior to 1985) and participation (yes/no) in weight training, stretching, or calisthenic exercises at least twice per week during the preceding 12 months. Statistical analyses Frequency tables were developed for all study variables and are presented by injury type for men and women separately. Polytomous logistic regression was used to assess the relation between injury status (activity related, “other,” and none) and each of the three exposures: the type and duration of physical activity and cardiorespiratory fitness level. The potential modifying effects of body mass index on the relation among physical activity, cardiorespiratory fitness, and injury status were assessed by the inclusion of interaction terms in the regression models using the dichotomous body mass index variable. The following variables were included in the final model as categorical variables: history of previous injury and participation in weight training, stretching, and calisthenic exercises. Age (years), cardiorespiratory fitness level (seconds), and body mass index were included in the final model as continuous variables. Statistical significance was set at the α = 0.05 level for all analyses. All analyses were conducted using SAS statistical software (SAS Institute, Inc., Cary, North Carolina). RESULTS The distribution of descriptive characteristics by injury type for men and women is shown in table 1. Among men, 16 percent reported an activity-related injury and 8 percent reported an “other” injury in the 12 months preceding the survey. Among women, 14 percent reported an activity-related injury and 9 percent reported an “other” injury. As age increased among men and women, the prevalence of activity-related injuries decreased, while the prevalence of “other” injuries was relatively stable among men. Among women aged 41–60 years, the prevalence of “other” injury was more than twice as high as among those aged 20–40 years and nearly twice as high among women over age 60. The proportion of men and women with activity-related injuries increased with increasing duration of physical activity and higher cardiorespiratory fitness levels. The majority of all subjects with either activity-related or “other” injuries reported having a previous injury. TABLE 1. Characteristics of subjects according to injury type and gender, Aerobics Center Longitudinal Study, 1970–1986 Characteristic  Injury type   Men (n = 4,034)   Women (n = 967)   Activity related   Other   None   Activity related   Other   None   No.  %  No.  %  No.  %  No.  %  No.  %  No.  %  Age (years)                           20–40  84  13.0  25  7.4  245  8.0  27  20.3  6  6.9  106  14.2   41–60  460  71.1  233  69.1  1,980  64.9  90  67.7  70  80.5  470  62.9   >60  103  15.9  79  23.4  825  27.0  16  12.0  11  12.6  171  22.9  Body mass index*                           Normal  346  53.5  143  42.4  1,388  45.5  110  82.7  66  75.9  620  83.0   Overweight/obese  301  46.5  194  57.6  1,662  54.5  23  17.3  21  24.1  127  17.0  PA† type                           Sedentary  111  17.2  132  39.2  1,172  38.4  36  27.1  41  47.1  327  43.8   Walker  44  6.8  32  9.5  307  10.1  27  20.3  13  14.9  164  22.0   Runner  416  64.3  135  40.1  1,305  42.8  59  44.4  25  28.7  210  28.1   Sport  76  11.7  39  11.6  266  8.7  11  8.3  8  9.2  46  6.2  PA duration                           Low  186  28.7  170  50.4  1,588  52.1  46  34.6  50  57.5  395  52.9   Moderate  227  35.1  98  29.1  932  30.6  40  30.1  21  24.1  218  29.2   High  234  36.2  69  20.5  530  17.4  47  35.3  16  18.4  134  17.9  Fitness level                           Low  24  3.7  56  16.6  422  13.8  11  8.3  16  18.4  110  14.7   Moderate  153  23.6  103  30.6  1,089  35.7  43  32.3  33  37.9  302  40.4   High  470  72.6  178  52.8  1,539  50.5  79  59.4  38  43.7  335  44.8  Previous injury                           Yes  503  77.8  236  70.0  1,691  55.4  86  64.7  54  62.1  283  37.9   No  144  22.3  101  30.0  1,359  44.6  47  35.3  33  37.9  464  62.1  Weight training                           Yes  396  61.2  268  79.5  2,397  78.6  95  71.4  76  87.4  664  88.9   No  251  38.8  69  20.5  653  20.8  38  28.6  11  12.6  83  11.1  Stretching                           Yes  263  40.6  202  59.9  1,942  63.7  58  43.6  55  63.2  462  61.8   No  384  59.4  135  40.1  1,108  36.3  75  56.4  32  36.8  285  3.7  Calisthenics                           Yes  471  72.8  283  84.0  2,439  80.0  88  66.2  70  80.5  585  78.3   No  176  27.2  54  16.0  611  20.0  45  33.8  17  19.5  162  21.7   Total  647    337    3,050    133    87    747    Characteristic  Injury type   Men (n = 4,034)   Women (n = 967)   Activity related   Other   None   Activity related   Other   None   No.  %  No.  %  No.  %  No.  %  No.  %  No.  %  Age (years)                           20–40  84  13.0  25  7.4  245  8.0  27  20.3  6  6.9  106  14.2   41–60  460  71.1  233  69.1  1,980  64.9  90  67.7  70  80.5  470  62.9   >60  103  15.9  79  23.4  825  27.0  16  12.0  11  12.6  171  22.9  Body mass index*                           Normal  346  53.5  143  42.4  1,388  45.5  110  82.7  66  75.9  620  83.0   Overweight/obese  301  46.5  194  57.6  1,662  54.5  23  17.3  21  24.1  127  17.0  PA† type                           Sedentary  111  17.2  132  39.2  1,172  38.4  36  27.1  41  47.1  327  43.8   Walker  44  6.8  32  9.5  307  10.1  27  20.3  13  14.9  164  22.0   Runner  416  64.3  135  40.1  1,305  42.8  59  44.4  25  28.7  210  28.1   Sport  76  11.7  39  11.6  266  8.7  11  8.3  8  9.2  46  6.2  PA duration                           Low  186  28.7  170  50.4  1,588  52.1  46  34.6  50  57.5  395  52.9   Moderate  227  35.1  98  29.1  932  30.6  40  30.1  21  24.1  218  29.2   High  234  36.2  69  20.5  530  17.4  47  35.3  16  18.4  134  17.9  Fitness level                           Low  24  3.7  56  16.6  422  13.8  11  8.3  16  18.4  110  14.7   Moderate  153  23.6  103  30.6  1,089  35.7  43  32.3  33  37.9  302  40.4   High  470  72.6  178  52.8  1,539  50.5  79  59.4  38  43.7  335  44.8  Previous injury                           Yes  503  77.8  236  70.0  1,691  55.4  86  64.7  54  62.1  283  37.9   No  144  22.3  101  30.0  1,359  44.6  47  35.3  33  37.9  464  62.1  Weight training                           Yes  396  61.2  268  79.5  2,397  78.6  95  71.4  76  87.4  664  88.9   No  251  38.8  69  20.5  653  20.8  38  28.6  11  12.6  83  11.1  Stretching                           Yes  263  40.6  202  59.9  1,942  63.7  58  43.6  55  63.2  462  61.8   No  384  59.4  135  40.1  1,108  36.3  75  56.4  32  36.8  285  3.7  Calisthenics                           Yes  471  72.8  283  84.0  2,439  80.0  88  66.2  70  80.5  585  78.3   No  176  27.2  54  16.0  611  20.0  45  33.8  17  19.5  162  21.7   Total  647    337    3,050    133    87    747    * Body mass index categories are defined as normal (<25 kg/m2) and overweight/obese (≥25 kg/m2). † PA, physical activity. View Large The distribution of activity-related and “other” injuries by body site among all subjects is presented in figure 1. Activity-related injuries were more common in the foot, ankle, lower leg, and thigh than “other” injuries. The proportion of “other” back and arm injuries was higher than the proportion of the activity-related back and arm injuries. The frequency of knee, hip, wrist, and hand injuries was similar for activity-related and “other” injuries. FIGURE 1. View largeDownload slide Distribution of activity-related and “other” injuries by body site among men and women, Aerobics Center Longitudinal Study, 1970–1986. FIGURE 1. View largeDownload slide Distribution of activity-related and “other” injuries by body site among men and women, Aerobics Center Longitudinal Study, 1970–1986. The independent associations between physical activity type, duration, and cardiorespiratory fitness level and musculoskeletal injury in men and women are shown in table 2. Body mass index did not modify the association between physical activity or cardiorespiratory fitness level and injury status and, therefore, regression models were adjusted for age, previous injury, body mass index, weight training, stretching, and calisthenic exercises. Running and sport participation were associated with approximately 2.3 times higher risk of activity-related injury in men and with 1.5–1.7 times higher risk in women. No significant difference was found between walkers and sedentary subjects for the risk of activity-related injury in women; male walkers had 73 percent higher risk of activity-related injury than did sedentary men. TABLE 2. Odds ratios and 95 percent confidence intervals for activity-related and other injury and for type and duration of physical activity and cardiorespiratory fitness among 5,001 men and women, Aerobics Center Longitudinal Study, 1970–1986   No.  Activity-related injury   Other injury     OR*,†  95% CI*  OR†  95% CI  Men             PA* type              Sedentary  1,415  1.00    1.00      Walker  383  1.73  1.18, 2.54  1.01  0.67, 1.54    Runner  1,856  2.38  1.87, 3.04  0.87  0.68, 1.11    Sport  381  2.26  1.62, 3.15  1.20  0.81, 1.77   PA duration              Low  1,944  1.00    1.00      Moderate  1,257  1.71  1.37, 2.13  0.98  0.74, 1.28    High  833  2.78  2.21, 3.50  1.18  0.86, 1.60   Fitness level              Low  502  1.00    1.00      Moderate  1,345  2.35  1.50, 3.70  0.70  0.50, 1.00    High  2,187  3.99  2.58, 6.18  0.84  0.60, 1.17  Women             PA type              Sedentary  404  1.00    1.00      Walker  204  1.35  0.77, 2.35  0.65  0.34, 1.27    Runner  294  1.68  1.03, 2.74  0.96  0.54, 1.70    Sport  65  1.50  0.68, 3.28  1.38  0.59, 3.25   PA duration              Low  491  1.00    1.00      Moderate  279  1.33  0.82, 2.14  0.79  0.45, 1.37    High  197  2.03  1.25, 3.30  0.87  0.46, 1.62   Fitness level              Low  137  1.00    1.00      Moderate  378  1.26  0.61, 2.58  0.79  0.41, 1.52    High  452  1.78  0.88, 3.58  0.85  0.44, 1.64    No.  Activity-related injury   Other injury     OR*,†  95% CI*  OR†  95% CI  Men             PA* type              Sedentary  1,415  1.00    1.00      Walker  383  1.73  1.18, 2.54  1.01  0.67, 1.54    Runner  1,856  2.38  1.87, 3.04  0.87  0.68, 1.11    Sport  381  2.26  1.62, 3.15  1.20  0.81, 1.77   PA duration              Low  1,944  1.00    1.00      Moderate  1,257  1.71  1.37, 2.13  0.98  0.74, 1.28    High  833  2.78  2.21, 3.50  1.18  0.86, 1.60   Fitness level              Low  502  1.00    1.00      Moderate  1,345  2.35  1.50, 3.70  0.70  0.50, 1.00    High  2,187  3.99  2.58, 6.18  0.84  0.60, 1.17  Women             PA type              Sedentary  404  1.00    1.00      Walker  204  1.35  0.77, 2.35  0.65  0.34, 1.27    Runner  294  1.68  1.03, 2.74  0.96  0.54, 1.70    Sport  65  1.50  0.68, 3.28  1.38  0.59, 3.25   PA duration              Low  491  1.00    1.00      Moderate  279  1.33  0.82, 2.14  0.79  0.45, 1.37    High  197  2.03  1.25, 3.30  0.87  0.46, 1.62   Fitness level              Low  137  1.00    1.00      Moderate  378  1.26  0.61, 2.58  0.79  0.41, 1.52    High  452  1.78  0.88, 3.58  0.85  0.44, 1.64  * OR, odds ratio; CI, confidence interval; PA, physical activity. † Each type, duration, and fitness models are separate models and each is adjusted for age, body mass index, previous injury, weight training, stretching, and calisthenic exercises. View Large In men with activity-related injuries, both moderate and high duration of physical activity per week and moderate and high cardiorespiratory fitness levels were associated with a significant increased risk of musculoskeletal injury compared with subjects in the low categories of duration and cardiorespiratory fitness. The risk of activity-related injury among men in the highest duration category was 2.8 times as high as for men in the low duration category. Highly fit men had the greatest risk of musculoskeletal injury, almost four times that of men in the lowest cardiorespiratory fitness category. Men with a moderate duration of physical activity per week (1.25–3.75 hours/week) and moderate cardiorespiratory fitness levels had about half the risk of activity-related injury as did those in the high duration and cardiorespiratory fitness categories. In women, the high level of duration and cardiorespiratory fitness was significantly associated with an increased risk of activity-related injury compared with women who had a low level of duration and cardiorespiratory fitness. The multivariate relation between activity-related injury and activity type, duration, and cardiorespiratory fitness in nonsedentary subjects (n = 2,931) is shown in table 3. Compared with that in walkers, the risk of activity-related injuries was not different for runners and sport participants for any duration of activity per week. Among walkers and sport participants, an increasing duration of activity per week did not significantly increase the risk of injury. In contrast, among runners the risk of injury increased with increasing duration of activity per week. TABLE 3. Odds ratios and 95 percent confidence intervals for the relation between activity-related injury and type of physical activity stratified by duration of physical activity and the relation between activity-related injury and duration of physical activity stratified by type of physical activity among nonsedentary subjects, Aerobics Center Longitudinal Study, 1970–1986   No.  Duration of PA*     <1.25 hours/week   1.25–3.75 hours/week   >3.75 hours/week     OR*,†  95% CI*  OR†  95% CI  OR†  95% CI  Type of PA                 Walker  542  1.00    1.00    1.00     Runner  1,990  0.45  0.20, 1.01  1.38  0.84, 2.27  1.40  0.87, 2.26   Sport  399   0.66   0.17, 2.48   1.15   0.62, 2.13   1.09   0.65, 1.83     No.  Type of PA     Walker   Runner   Sport     OR†   95% CI*   OR†   95% CI   OR†   95% CI   Duration of PA                 <1.25 hours/week  569  1.00    1.00    1.00     1.25–3.75 hours/week  1,417  0.66  0.31, 1.43  1.47  1.08, 2.00  0.91  0.27, 3.05   >3.75 hours/week  945  1.16  0.54, 2.52  2.38  1.72, 3.29  1.30  0.40, 4.23    No.  Duration of PA*     <1.25 hours/week   1.25–3.75 hours/week   >3.75 hours/week     OR*,†  95% CI*  OR†  95% CI  OR†  95% CI  Type of PA                 Walker  542  1.00    1.00    1.00     Runner  1,990  0.45  0.20, 1.01  1.38  0.84, 2.27  1.40  0.87, 2.26   Sport  399   0.66   0.17, 2.48   1.15   0.62, 2.13   1.09   0.65, 1.83     No.  Type of PA     Walker   Runner   Sport     OR†   95% CI*   OR†   95% CI   OR†   95% CI   Duration of PA                 <1.25 hours/week  569  1.00    1.00    1.00     1.25–3.75 hours/week  1,417  0.66  0.31, 1.43  1.47  1.08, 2.00  0.91  0.27, 3.05   >3.75 hours/week  945  1.16  0.54, 2.52  2.38  1.72, 3.29  1.30  0.40, 4.23  * PA, physical activity; OR, odds ratio; CI, confidence interval. † Adjusted for cardiorespiratory fitness, body mass index, age, previous injury, weight training, stretching, and calisthenic exercises. View Large DISCUSSION This is the first report of the risk of musculoskeletal injury associated with selected types and duration of physical activity and cardiorespiratory fitness among recreationally active adults. The risk of activity-related injury among adults increased for runners, sport participants, persons engaging in more than 1.25 hours per week of physical activity, and persons with moderate to high cardiorespiratory fitness levels. Walking as a primary mode of physical activity was not associated with a significant increase in risk of injury among women, and injury risk did not increase with longer durations of walking per week. The foot and ankle were the most frequent sites of activity-related injuries while the back was the most common site of “other” injury. Musculoskeletal injuries among physically active populations tend to occur in the lower extremity (7, 9–19). Activity-related back injuries, however, tend to be sport specific, because sports such as gymnastics, golf, and weight lifting have a higher incidence of back injuries than other sports (28, 29). Population-based and health club studies describing the distribution of general musculoskeletal injuries among adults also report the back as one of the top three sites of injury (30, 31). In this study we measured four of five components of physical activity that can contribute to the risk of musculoskeletal injury: cardiorespiratory fitness and type, frequency, and duration of activity. We do not have a measure of the intensity of exercise for all subjects, primarily those reporting participation in sports activities such as swimming, cycling, and other strenuous sports. Our findings regarding high cardiorespiratory fitness level as a risk factor for injury contrast with results from military studies reporting low cardiorespiratory fitness levels being associated with increased training injuries (13, 20, 21). One explanation may be that the cardiorespiratory fitness level in our study is acting as a surrogate for activity intensity or some other unmeasured component of activity. However, among nonsedentary subjects, a higher cardiorespiratory fitness level was still associated with an increased risk of activity-related injury, even after adjusting for the type and duration of physical activity. Furthermore, among just the runners and walkers, cardiorespiratory fitness was still associated with a nonsignificant 50–74 percent increased risk of injury after controlling for type (running vs. walking), duration (hours per week of activity), and intensity (average pace). Even with these findings, these results suggest that other factors, such as biomechanical or structural abnormalities or personality traits, may also play a role in the relation between physical activity and the risk of musculoskeletal injuries. Differences in the study design and sample between our study and military studies may also explain our contrasting findings regarding cardiorespiratory fitness as a risk factor for injury. The military population is young (18–23 years old), while our subjects range from 20 to 85 years. Being older on average, our subjects may have sustained more injuries over their lifetime than younger persons, thereby increasing their risk of subsequent injuries. We controlled for previous injuries. However, some subjects may not have reported injuries that occurred a long time in the past. In military studies, cardiorespiratory fitness was measured immediately preceding entrance into a vigorous basic training program, and subjects most likely did not have sufficient exercise experience prior to testing to have high cardiorespiratory fitness levels. Because their lifetime exercise experience is higher, our subjects' baseline cardiorespiratory fitness levels may have been higher than those of military trainees. Cardiorespiratory fitness is usually measured by 2-mile-run times in military studies, and therefore we cannot directly compare our baseline cardiorespiratory fitness measure (seconds on treadmill) with theirs. It would be interesting to know how cardiorespiratory fitness measured after a 6-week basic military training program, when subjects have accommodated to the increased activity levels, correlates with injuries reported over the next 12 months. When measured over a year as in our study, it may be that even military trainees with high cardiorespiratory fitness levels will be at high risk of injury. Military trainees engage in a well-controlled physical training program, and training volume is essentially equal among subjects. Our subjects are free to vary their activity level and therefore have a wide range of activity time per week. One military study that did report some variation in hours per week of training found, similar to ours, that training time per week was correlated with injuries (32). In addition, military studies use a stricter definition of injury requiring that an injury must have required medical attention and be recorded in the medical record. We included any injury the subject attributed to physical activity. In summary, the association of a higher level of cardiorespiratory fitness with higher injury rates in this cohort most likely reflects our subjects' high lifetime activity levels. A higher level of cardiorespiratory fitness may also be acting as a surrogate for some unmeasured risk factor for injury such as biomechanical or structural abnormalities, personality traits, or exercise behavior patterns. Identifying activities that provide an adequate amount of physical activity to improve health but have a lower risk of injury is an important public health goal. Walking is a popular form of moderate physical activity; an estimated 138 million US adults report walking for exercise (33). According to recent national estimates, 1.4 percent of US adults are injured while walking for exercise (33). Several exercise training studies report a lower frequency of injury during walking activities than during jogging (34–36). In an epidemiologic study comparing runners and walkers, walkers had a lower risk of activity-related injury (37). In addition, the injury risk among runners significantly increased with increasing duration of running per day, while increased duration of walking per day did not significantly increase the injury risk among walkers. We report similar findings that, among walkers and sport participants, the risk of injury did not increase with increased duration of activity per week. In contrast, among runners the risk of activity-related injury increases with longer duration of physical activity per week in a classic dose-response pattern. None of the relations among physical activity, cardiorespiratory fitness level, and the risk of “other” injuries was statistically significant, which is likely due to the small number of “other” injuries reported (approximately 8 percent). However, a consistent trend is evident, suggesting that the type of physical activity and cardiorespiratory fitness level may provide some protection against an “other” injury. Running decreased the risk of an “other” injury 13 percent among men, and walking decreased the risk of “other” injury 35 percent among women. Moderate cardiorespiratory fitness also decreased the risk of an “other” injury 30 percent among men and 21 percent among women. A strong and efficient neuromuscular system is better able to adapt to acute stresses and subsequently may be more resistant to physical injury (38–41). Thus, moderate physical activity in amounts to achieve a moderate level of cardiorespiratory fitness may provide an adequate stimulus to strengthen the neuromuscular and skeletal systems and decrease the risk of injury, regardless of the cause of injury. In addition to the small number of “other” injuries reported, several other limitations of these data should be mentioned. First, inadequate recall of information used to classify each subject into exposure and outcome groups may introduce measurement error (42). Measurement error in this sample is expected to be low because the accuracy of recall of self-reported health information among subjects in the Aerobics Center Longitudinal Study has been previously reported as excellent (98 percent sensitivity, 99 percent specificity) (1). Second, subjects reported if an injury was related to participation in a run/walk/jog program or a strenuous sport program or due to some “other” cause. The cause of injuries reported in the “other” category was not specified. Therefore, some injuries due to sports activities, such as golf, may not have been perceived by subjects to fit into the run/walk/jog or strenuous sports categories. These other types of sports injuries may have gone unreported or may have been included in the “other” category and, therefore, it is possible that some activity-related injuries may have been classified as other injuries. This type of misclassification would most likely be nondifferential and would bias the risk estimate toward the null. Third, 8 percent of sedentary subjects reported an activity-related injury. The sedentary category included subjects who did not report any participation in the specific activities listed on the survey. Therefore, sedentary subjects did not consider themselves to be regularly physically active but may have been injured during occasional participation in sport activities on weekends or at social gatherings. Lastly, cardiorespiratory fitness levels may change significantly from baseline to follow-up, an average of 8.2 years in this sample. However, a unique feature of this cohort is that cardiorespiratory fitness levels have been reported to be relatively constant over time. Using subjects from the same cohort, Blair et al. (43) report that approximately 82 percent of subjects remained fit from the baseline treadmill examination to a second treadmill examination (the mean time between examinations was 5 years). In our sample, using body mass index as a proxy for cardiorespiratory fitness, over 90 percent of subjects maintained stable weight from baseline to follow-up. An additional limitation is that, although the response rate was adequate, differences may exist between the subjects who responded to the follow-up survey and those who did not respond. Macera et al. (44) conducted a comparison of subjects who did and did not return the 1982 follow-up survey in this same cohort. Using baseline laboratory measurements and a personal health history, nonrespondents and respondents were equally healthy. However, participants reporting more positive health behaviors and more family history of chronic disease were more likely to respond to the mail survey. Assuming that the response patterns reported by Macera et al. are similar to those of the current study using the 1986 follow-up survey, subjects included in this sample most likely represent health conscious adults who demonstrate behaviors consistent with good health, such as engaging in physical activity and not smoking. Thus, the results of this study may not be applicable to all segments of the population. Several aspects of these data and the statistical procedures contribute to the strengths of this study. Detailed data on physical activity, cardiorespiratory fitness, and injury were obtained from a large sample of adult men and women. Subjects reported a wide range of activity levels, including no regular physical activity, which allowed comparison of injury experiences across different levels and duration of activity. Cardiorespiratory fitness was directly measured by a treadmill graded exercise test and serves as an objective measurement of cardiorespiratory fitness level. Simultaneously estimating the association between each exposure and activity-related and “other” injuries using polytomous logistic regression allowed for direct comparison of the risk estimates for each injury cause (45). It is well known that subgroups of vigorously active populations such as runners, scholastic athletes, and military trainees experience high rates of musculoskeletal injury (7, 9–19, 21). The results of this study indicate that the moderate types and duration of physical activity promoted by private and federal health organizations have lower injury risk than more vigorous types and longer durations of physical activity. Walking for exercise was not associated with a significant increased risk of activity-related injury, even among those walkers with the highest duration of activity per week. This suggests that for most adults, walking is a safe form of physical activity. However, to minimize the risk of injury, public health practitioners need to target injury prevention messages to adults engaged in vigorous types of physical activity. Reprint requests to Dr. Jennifer M. Hootman, Centers for Disease Control and Prevention, 4770 Buford Highway N.E., MS K-45, Atlanta, GA 30341 (e-mail: [email protected]). This study was supported by grant NIH NIA AG06945 from the National Institutes of Health awarded to the Cooper Institute for Aerobics Research. The authors thank Dorothy Davis of the University of South Carolina Prevention Research Center and the staff of the Cooper Institute for Aerobics Research for collecting and managing the data used in this study. REFERENCES 1. Blair S, Goodyear N, Gibbons L, et al. Physical fitness and incidence of hypertension in healthy normotensive men and women. JAMA  1984; 252: 487–90. Google Scholar 2. Blair S, Kohl H, Paffenbarger R, et al. Physical fitness and all-cause mortality: a prospective study of healthy men and women. JAMA  1989; 262: 2395–401. Google Scholar 3. US Department of Health and Human Services. Physical activity and health: a report of the Surgeon General. Atlanta, GA: Centers for Disease Control and Prevention, 1996. Google Scholar 4. American College of Sports Medicine. The recommended quantity and quality of exercise for developing and maintaining cardiorespiratory and muscular fitness and flexibility in healthy adults. Med Sci Sports Exerc  1998; 30: 975–91. Google Scholar 5. Pate R, Pratt M, Blair S, et al. Physical activity and public health: a recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine. JAMA  1995; 273: 402–7. Google Scholar 6. Uitenbroek D. Sports, exercise, and other causes of injuries: results of a population study. Res Q Exerc Sport  1996; 67: 380–5. Google Scholar 7. Koplan J, Powell K, Sikes R, et al. An epidemiologic study of the benefits and risks of running. JAMA  1982; 248: 3118–21. Google Scholar 8. Koplan J, Siscovick D, Goldbaum G. The risks of exercise: a public health view of injuries and hazards. Public Health Rep  1985; 100: 189–95. Google Scholar 9. Bovens A, Janssen G, Vermeer H, et al. Occurence of running injuries in adults following a supervised training program. Int J Sports Med  1989; 10(suppl): S186–90. Google Scholar 10. Caselli M, Longobardi S. Lower extremity injuries at the New York City marathon. J Am Podiatr Med Assoc  1997; 87: 34–7. Google Scholar 11. Blair S, Kohl H, Goodyear N. Rates and risks for running and exercise injuries: studies in three populations. Res Q Exerc Sport  1987; 58: 221–8. Google Scholar 12. Jacobs S, Berson B. Injuries to runners: a study of entrants to a 10,000 meter race. Am J Sports Med  1986; 14: 151–5. Google Scholar 13. Jones BH, Bovee MW, Harris J, et al. Intrinsic risk factors for exercise-related injuries among male and female Army trainees. Am J Sports Med  1993; 21: 705–10. Google Scholar 14. Lysholm J, Wiklander J. Injuries in runners. Am J Sports Med  1987; 15: 168–71. Google Scholar 15. Macera C, Pate R, Powell K, et al. Predicting lower-extremity injuries among habitual runners. Arch Intern Med  1989; 149: 2565–8. Google Scholar 16. Macera C, Pate R, Woods J, et al. Postrace morbidity among runners. Am J Prev Med  1991; 7: 194–8. Google Scholar 17. Marti B, Vader J, Minder C, et al. On the epidemiology of running injuries: the 1984 Bern Grand-Prix Study. Am J Sports Med  1988; 16: 285–94. Google Scholar 18. Marti B. Benefits and risks of running among women: an epidemiologic study. Int J Sports Med  1988; 9: 92–8. Google Scholar 19. Walter S, Hart L, McIntosh J. The Ontario cohort study running-related injuries. Arch Intern Med  1989; 149: 2561–4. Google Scholar 20. Shaffer RA, Brodine SK, Almeida SA, et al. Use of measures of physical activity to predict stress fractures in young men undergoing a rigorous physical training program. Am J Epidemiol  1999; 149: 236–42. Google Scholar 21. Knapik J, Ang P, Reynolds K, et al. Physical fitness, age, and injury incidence in infantry soldiers. J Med  1993; 35: 598–603. Google Scholar 22. Balke B, Ware R. An experimental study of physical fitness on Air Force personnel. US Armed Forces Med J  1959; 10: 675–88. Google Scholar 23. Cooper Institute for Aerobics Research. Procedures manual for Cooper Clinic technicians. Dallas, TX: Cooper Clinic, 1986. Google Scholar 24. Pollock M, Bohannon R, Cooper K, et al. A comparative analysis of four protocols for maximal treadmill stress testing. Am Heart J  1976; 92: 39–46. Google Scholar 25. Pollock M, Foster C, Schmidt D, et al. Comparative analysis of physiologic responses to three different maximal graded exercise test protocols in healthy women. Am Heart J  1982; 103: 363–73. Google Scholar 26. Lee CD, Blair SN, Jackson AS. Cardiorespiratory fitness, body composition, and all-cause and cardiovascular disease mortality in men. Am J Clin Nutr  1999; 69: 373–80. Google Scholar 27. Blair S, Kohl H, Barlow C, et al. Changes in physical fitness and all-cause mortality: a prospective study of healthy and unhealthy men. JAMA  1995; 273: 1093–8. Google Scholar 28. Caine D, Caine C, Linder K. The epidemiology of sports injuries. Champaign, IL: Human Kinetics Press, 1996. Google Scholar 29. Wadley GH, Albright JP. Women's intercollegiate gymnastics: injury patterns and “permanent” medical disability. Am J Sports Med  1993; 21: 314–20. Google Scholar 30. vanMechelen W, Twisk J, Molendijk A, et al. Subject-related risk factors for sports injuries: a 1-year prospective study in young adults. Med Sci Sports Exerc  1996; 28: 1171–9. Google Scholar 31. Sedgwick A, Smith D, Davies M. Musculoskeletal status of men and women who entered a fitness program. Med J Aust  1988; 148: 385–91. Google Scholar 32. Almeida SA, Williams, KM, Shaffer RA, et al. Epidemiological patterns of musculoskeletal injuries and physical training. Med Sci Sports Exerc  1999; 31: 1176–82. Google Scholar 33. Powell K, Heath G, Kresnow M, et al. Injury rates from walking, gardening, weightlifting, outdoor bicycling, and aerobics. Med Sci Sports Exerc  1998; 30: 1246–9. Google Scholar 34. Pollock M, Carroll J, Graves J, et al. Injuries and adherence to walk/jog and resistance training programs in the elderly. Med Sci Sports Exerc  1991; 23: 1194–200. Google Scholar 35. Suter E, Marti B, Gutzwiller F. Jogging or walking—comparison of health effects. Ann Epidemiol  1994; 4: 375–81. Google Scholar 36. Carroll J, Pollock M, Graves J, et al. Incidence of injury during moderate and high-intensity walking training in the elderly. J Gerontol  1992; 47: M61–6. Google Scholar 37. Colbert LH, Hootman JM, Macera CA. Physical activity-related injuries in walkers and runners in the Aerobics Center Longitudinal Study. Clin J Sport Med  2000; 10: 259–63. Google Scholar 38. Radin E. Role of muscles in protecting athletes from injury. Acta Med Scand  1986; 711: 143–7. Google Scholar 39. Sale D. Neural adaptation to resistance training. Med Sci Sports Exerc  1988; 20(suppl): S135–45. Google Scholar 40. Vuori I. Exercise and physical health: musculoskeletal health and functional capabilities. Res Q Exerc Sport  1995; 66: 276–85. Google Scholar 41. Wolfson L, Judge J, Whipple R, et al. Strength is a major factor in balance, gait, and the occurrence of falls. J Gerontol  1995; 50A: 64–7. Google Scholar 42. Gordis L. Epidemiology. Philadelphia, PA: WB Saunders Company, 1996. Google Scholar 43. Blair SN, Kohl HW, Barlow CE, et al. Changes in physical fitness and all-cause mortality: a prospective study of healthy and unhealthy men. JAMA  1995; 273: 1093–8. Google Scholar 44. Macera CA, Jackson K, Davis D, et al. Patterns of non-response to a mail survey. J Clin Epidemiol  1990; 43: 1427–30. Google Scholar 45. Hosmer D, Lemeshow S. Applied logistic regression. New York, NY: John Wiley & Sons, Inc, 1989. Google Scholar

Journal

American Journal of Epidemiology – Oxford University Press

Published: Aug 1, 2001

There are no references for this article.